CN113397530B

CN113397530B - Intelligent correction system and method capable of evaluating knee joint function

Info

Publication number: CN113397530B
Application number: CN202110668711.6A
Authority: CN
Inventors: 闫琪; 屠一峰; 左伊鹏; 黄丽; 纪豪
Original assignee: CHINA INSTITUTE OF SPORT SCIENCE
Current assignee: CHINA INSTITUTE OF SPORT SCIENCE
Priority date: 2021-06-16
Filing date: 2021-06-16
Publication date: 2022-03-18
Anticipated expiration: 2041-06-16
Also published as: CN113397530A

Abstract

The invention discloses an intelligent correction system and a method capable of performing knee joint function evaluation, wherein the intelligent correction system comprises: the image acquisition equipment and the evaluation and correction integrated machine comprise a display screen, a shell support and an intelligent analysis system, wherein the display screen is embedded on the surface of the shell, the shell is fixedly connected with the shell support, the intelligent analysis system is connected with the display screen, the image acquisition equipment and the correction integrated machine are arranged in parallel, the image acquisition equipment and the correction integrated machine are arranged perpendicular to the ground, and the intelligent analysis system is packaged in the shell. The invention realizes the integration of knee joint function evaluation and correction, improves the evaluation accuracy, quantifies key indexes, can carry out specialized evaluation and scientific exercise scheme exercise at any time and any place, and reduces the labor cost.

Description

Intelligent correction system and method capable of evaluating knee joint function

Technical Field

The invention relates to the technical field of computer vision capture, in particular to an intelligent correction system and method capable of performing knee joint function assessment.

Background

Through completing a series of test actions, knee joint dysfunction of a subject is discovered, and then targeted training is carried out to help a restricted person to correct existing problems, reduce the risk of sports injury and improve functions, which plays an important role for all sports participants. The traditional knee joint dysfunction detection method depends on visual observation (influencing test precision), depends on professional ability of an evaluator and cannot form a quantitative evaluation system.

The existing Human body action posture recognition based on the computer vision technology, such as a Human Trak motion analysis system and a digital FMS evaluation and training system, needs a user to wear a corresponding sensor, and certain influence is caused on the user to complete the action, so that the analysis result is inaccurate. Moreover, international human body motion analysis systems are expensive, evaluation indexes do not conform to the constitution of Chinese people, and domestic important data (such as training data of soldiers and high-level athletes) are easy to flow abroad, so that data information safety is influenced.

Disclosure of Invention

The invention aims to provide an intelligent correction system and method capable of performing knee joint function evaluation, which can be used for solving the problems in the prior art, can realize the acquisition of user action information by directly shooting the action of a user by using a camera without wearing any large or miniature sensor, and can improve the scientificity and the practicability of a detection method and the specialty of motion analysis by carefully screening and formulating a knee joint dysfunction evaluation method, indexes and a motion scheme, thereby playing the individuation and the digitization of physical training, and realizing the scientific training of athletes and the scientific body building of the ordinary people.

In order to achieve the purpose, the invention provides the following scheme: the invention provides an intelligent correction system capable of performing knee joint function evaluation, which comprises image acquisition equipment and an evaluation and correction all-in-one machine, wherein the evaluation and correction all-in-one machine comprises a display screen, a shell bracket and an intelligent analysis system,

the image acquisition equipment is used for acquiring motion images of the subject;

the display screen is used for displaying the interactive information between the subject and the intelligent correction system capable of performing knee joint function assessment;

the intelligent analysis system is used for analyzing the motion image of the subject to provide a personalized motion scheme;

the display screen inlays the dress and is in the casing surface, the casing with casing support fixed connection, intelligent analysis system with the display screen is connected, image acquisition equipment with correct all-in-one parallel arrangement, just image acquisition equipment correct all-in-one and all set up with ground vertically, intelligent analysis system encapsulates in the casing.

Preferably, the intelligent analysis system comprises a control module, a data storage module, an information processing module and a project management module,

the control module is used for issuing a control instruction;

the data storage module is used for storing the motion image of the subject;

the information processing module is used for identifying the action posture of the subject, evaluating the knee joint dysfunction, judging the action standard degree and making evaluation;

the project management module is used for configuring evaluation actions and training actions, forming an evaluation action scheme and a motion scheme and adjusting motion projects;

the control module is respectively connected with the information processing module and the project management module, the information processing module is connected with the data storage module, and the control module is respectively connected with the display screen and the image acquisition equipment.

Preferably, the action postures of the testee comprise knee joint mobility action postures, knee joint static standing posture postures and knee joint function action postures.

Preferably, the knee joint range of motion action includes left and right hip flexion straight leg lifting, left and right hip lateral lying abduction, left and right hip prone internal rotation, left and right knee joint flexion, left and right kneeling ankle dorsiflexion.

Preferably, the functional actions include deep squat, hip joint hinge, left and right single leg squat, left and right bow step squat, jump to the ground.

Preferably, the display screen adopts an LCD liquid crystal touch screen, and the LCD liquid crystal touch screen supports a 4K display effect.

Preferably, the control module employs an english viad Jetson TX2 core board.

Preferably, the intelligent correction system capable of performing knee joint function evaluation further comprises a floor rubber mat, the evaluation and correction all-in-one machine and the image acquisition equipment are fixedly arranged above the floor rubber mat, and the evaluation and correction all-in-one machine and the image acquisition equipment are perpendicular to the floor rubber mat.

Also provided is an intelligent correction method capable of performing knee joint function assessment, comprising the following steps:

the image acquisition equipment acquires the motion image of the subject and transmits the motion image to the evaluation and correction all-in-one machine,

based on the action image, the evaluation and correction all-in-one machine identifies the action posture of the knee joint and collects the action pain degree information of the testee;

based on the knee joint action posture and the action pain degree information, the evaluation and correction all-in-one machine evaluates the knee joint dysfunction of the testee and acquires the health state of the knee joint;

analyzing the health state of the knee joint and providing an individualized exercise scheme;

and monitoring the execution of the personalized motion scheme, proposing an execution evaluation result and correcting the action deviation of the subject.

Preferably, the personalized motion profile comprises a lower limb muscle strength exercise, a core stability exercise, a lower limb functional pattern exercise, a breathing exercise, a joint range of motion exercise and a lower limb joint stability exercise.

The invention discloses the following technical effects:

the intelligent correction system and the method for evaluating the knee joint function, which are provided by the invention, utilize the computer vision technology and the artificial intelligence technology to realize the intelligent evaluation and correction of the knee joint function, accurately and intelligently identify the knee joint function condition of a user in real time, improve the evaluation accuracy and quantify key indexes, and the system can also intelligently recommend a correction training scheme according to the problems obtained by the evaluation, thereby achieving a whole set of health services of detection, evaluation and correction, realizing the specialized evaluation and scientific knee joint function exercise at any time and any place, reducing the labor cost and facilitating the social publicity and the popularization of the scientific concept of the function exercise.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.

FIG. 1 is a schematic diagram of an intelligent orthotic system for evaluating knee joint function according to the present invention;

FIG. 2 is a schematic flow chart of an intelligent correction method for evaluating knee joint function according to the present invention;

FIG. 3 is a schematic diagram of a knee joint dysfunction assessment project according to an embodiment of the present invention;

FIG. 4 is a schematic diagram of a human skeleton based on computer vision according to an embodiment of the present invention;

FIG. 5 is a schematic diagram of a skeleton sequence identification process according to an embodiment of the present invention;

FIG. 6 is a schematic view of a personalized knee joint movement protocol according to the present invention;

fig. 7 is a schematic view of an image acquisition and processing process for evaluating knee joint dysfunction according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.

The invention provides an intelligent correction system and a method capable of evaluating knee joint function, the intelligent correction system is targeted at people with knee joint function degradation, many people in the current society have knee joint dysfunction, is common among the old people, the people with sedentary immobility and the people with excessive movement, is represented by the basic lower limb action modes of limited lower limb joint mobility, insufficient stability, incapability of completing squatting, arrow walking and the like, and even has slight chronic pain (the pain VAS score is less than 3 points), although the knee joints of the people do not reach the disease treatment degree, the daily life quality and the motor ability are affected, the people group is the main application object of the intelligent correction system, that is, the knee joint dysfunction referred to in this embodiment is considered to be a problem arising from the functional action point of view, and is not structural damage or a clear pathologically induced knee joint dysfunction. Referring to fig. 1-2, the intelligent correction system comprises an image acquisition device and an evaluation and correction integrated machine, the evaluation and correction integrated machine comprises a display screen, a shell support and an intelligent analysis system, the display screen is installed on the surface of the shell in an embedding mode, and the shell is fixedly connected with the shell support. Inside intelligent analysis system encapsulated the casing, intelligent analysis system was connected with the display screen, in this embodiment, the display screen adopted LCD liquid crystal touch screen, and LCD liquid crystal touch screen can support 4K display effect, guarantees the high definition demand of display content, makes that the experimenter can be clear sees the action detail. The control module employs an English WEDAJetson TX2 core board. The image acquisition device uses a 120 (H) 90 (V) wide-angle undistorted monocular camera. The image acquisition equipment and the correction all-in-one machine are arranged in parallel, and the image acquisition equipment and the correction all-in-one machine are arranged perpendicular to the ground. In order to ensure the accuracy of motion acquisition of the testee, the correction all-in-one machine and the image acquisition equipment are vertically arranged by taking the testee as the center.

The intelligent analysis system comprises a control module, a data storage module, an information processing module and a project management module. The examinee selects the evaluation item of the knee joint dysfunction by touching the LCD touch screen, the control module receives the evaluation item information, the display screen displays the evaluation item action list, the examinee selects the corresponding evaluation item action, the image acquisition device starts to acquire the action image of the examinee according to the starting instruction, and the image is transmitted to the data storage module to be stored. As shown in fig. 3, the evaluation items of knee joint dysfunction include: (1) the joint mobility detection needs to detect 10 actions, namely left and right hip flexion and straight leg elevation, left and right hip lateral lying and abduction, left and right hip prone inward rotation, left and right knee joint flexion and left and right kneeling ankle dorsiflexion. The evaluation finds out whether the joint activity is limited, and the problems of tension of soft tissues such as related muscles and ligaments or pain excitation points exist. (2) And (4) evaluating the body posture, wherein the static standing posture is the basic standing posture of the human body, and detecting whether the pelvis has heeling, rotation and poor knee joint posture (knee eversion, knee varus, knee flexion and knee hyperextension). (3) Functional action evaluation needs to evaluate 7 actions, namely deep squat, hip joint hinge, left-right single-leg squat, left-right bow step squat and jump landing, and the evaluation finds out whether the problems of poor functional action mode, insufficient core control capacity, limited hip bending and ankle dorsiflexion capacity, relevant muscle group tension, poor proprioception and the like exist. After each evaluation action is finished, the subject selects the pain degree according to the self condition and the prompt of the display screen, and the pain degree information is transmitted to the data storage module.

The data storage module establishes a basic database of the collected evaluation images of the knee joint dysfunction of the subject, wherein the basic database comprises a positive sample and a negative sample. And transmitting the knee joint dysfunction evaluation image of the testee to an information processing module for posture recognition of the testee. The information processing module identifies a pose of the subject based on computer vision. Computer vision human motion recognition functions to evaluate key points of a human skeleton, such as joint points (joints) and bones (bones), the joint points function to link two adjacent bones. Therefore, the skeleton is simplified into an image (graph) composed of points and edges. The points correspond to the joint points in the skeleton, and the edges correspond to the skeleton in the skeleton, as shown in fig. 4.

In the embodiment, the information processing module applies various human body recognition related neural network model algorithms. As shown in fig. 5, after the evaluation image of the knee joint dysfunction of the subject is processed by the posture recognition and posture tracking algorithm, a discrete sequence of key points (joint points) 2d or 3d with respect to the time dimension is obtained. According to the time dimension key point (joint point) 2d or 3d discrete sequence, the information processing module continues to perform motion recognition algorithm analysis on displacement and angles of the joint point and the skeleton of a skeleton sequence, so that multi-dimensional parameter index evaluation and behavior action recognition results of the skeleton of a body can be obtained, a knee joint dysfunction evaluation project is combined, action postures and action pain degrees of the knee joint dysfunction evaluation project of a subject are recognized, a knee joint health state is obtained, and the knee joint health state is transmitted to the project management module. Knee joint health includes the health of bones and muscles.

The project management module provides a personalized exercise scheme according to the health condition of the knee joint, as shown in fig. 6, and sends the personalized exercise scheme to the control module, and the personalized exercise scheme is transmitted to the display screen by the control module for display. The subject selects a correction exercise item according to the content of the personalized exercise scheme, and performs knee joint dyskinesia correction according to the provided action essentials and standard action demonstration. The image acquisition equipment acquires a moving image of a subject and sends the moving image to the information storage module, the information storage module sends the moving image to the information processing module, the information processing module identifies the motion posture of the subject, compares an identification result with a standard motion, judges the motion standard degree and gives a motion completion score and a motion deviation correction suggestion, and invalid exercise and even further damage are prevented.

The project management module can configure evaluation actions and training actions, configure action grouping, editing or deleting actions and can perform content editing such as action names, action descriptions, detection indexes, action videos and the like.

In addition, the project management module carries out dynamic management on the evaluation project, including the operation and use state of the existing evaluation project, whether the evaluation project is started or stopped; the project management module can also be configured with a plurality of versions and a plurality of action groups for managing the evaluation actions and the correction training actions of different versions, and can carry out the operations of adding, modifying and deleting.

The project management module configures a conclusion according to the configured index range for each detection action, each evaluation project and each correction training action at regular intervals; the conclusion is divided into action class, item class, body part class and correction training class, and the operations of adding, modifying and deleting can be carried out.

Scoring is performed in three dimensions, including test action scores, overall item assessment scores, body part scores, and index and weight configurations.

Further optimize the scheme, for guaranteeing image acquisition's comprehensiveness and clarity in this embodiment, correction system still includes the floor cushion in this embodiment, and the floor cushion is laid subaerially, and image acquisition equipment and evaluation correction all-in-one set are on the floor cushion to image acquisition equipment and evaluation correction all-in-one are perpendicular with the floor cushion. The position of the subject can be more clearly shown, and the image acquired at the position is more comprehensive and clear.

The image acquisition processing process in the evaluation of the invention is shown in fig. 7, firstly video data is acquired through a camera, a basic database is established, wherein the basic database comprises a positive example sample and a negative example sample, the characteristics of human bodies (namely human body joint point information) in the positive example sample and the negative example sample are respectively extracted, a positive example characteristic set and a negative example characteristic set are established, and the positive example characteristic set and the negative example characteristic set are transmitted to an information processing module; then, the information processing module extracts human body features (human body joint points, hereinafter referred to as joint points) such as a head, a left wrist, a right wrist, a left shoulder and the like in the video through a convolutional neural network, the data is acquired at 30 frames per second, and the human body joint point identification accuracy rate is over 93%. And finally, obtaining a corresponding evaluation result by combining a knee joint dysfunction evaluation method and indexes.

The intelligent correction system and the intelligent correction method capable of evaluating the knee joint function can realize the acquisition of the action information of the person to be evaluated without wearing any large or miniature sensor by the person to be evaluated, and the action of the person to be evaluated is shot by utilizing the high-definition camera, and the evaluation method, the index and the personalized movement scheme of the knee joint function disorder are elaborately screened and formulated by experts of the national sports institute sports science institute, so that the scientificity, the practicability, the specialty and the pertinence of the movement analysis are improved, the individuation and the digitization of the physical training are realized, the system is convenient to install, the use field is less limited, and various requirements of scientific training of athletes and scientific body building of common people can be met. Through practical application, the user obtains good exercise effect.

In the description of the present invention, it is to be understood that the terms "longitudinal", "lateral", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like, indicate orientations or positional relationships based on those shown in the drawings, are merely for convenience of description of the present invention, and do not indicate or imply that the referenced devices or elements must have a particular orientation, be constructed and operated in a particular orientation, and thus, are not to be construed as limiting the present invention.

The above-described embodiments are merely illustrative of the preferred embodiments of the present invention, and do not limit the scope of the present invention, and various modifications and improvements of the technical solutions of the present invention can be made by those skilled in the art without departing from the spirit of the present invention, and the technical solutions of the present invention are within the scope of the present invention defined by the claims.

Claims

1. An intelligent orthotic system that can perform knee joint functional assessment, comprising: comprises an image acquisition device and an evaluation and correction integrated machine, wherein the evaluation and correction integrated machine comprises a display screen, a shell bracket and an intelligent analysis system,

the display screen is used for displaying interaction information between the subject and the intelligent correction system capable of performing knee joint function evaluation and receiving feedback information of the subject;

the intelligent analysis system is used for analyzing the motion image of the subject and providing a personalized motion scheme, and comprises a control module, a data storage module, an information processing module and a project management module,

the control module is used for issuing a control instruction;

the data storage module is used for storing the motion image of the subject;

the information processing module is used for identifying the action postures of the testee, evaluating the functional disorder of the knee joint by combining the action pain degree, judging the action standard degree and evaluating, wherein the action postures of the testee comprise knee joint activity degree action postures, knee joint static standing posture postures and knee joint functional action postures, the knee joint activity degree actions comprise left and right hip flexion and straight leg lifting, left and right hip lateral lying abduction, left and right hip prone internal rotation, left and right knee joint flexion and left and right kneeling ankle dorsiflexion, and the functional actions comprise deep squatting, hip joint hinges, left and right single leg squatting, left and right bow step squatting and jumping and landing;

the control module is respectively connected with the information processing module and the project management module, the information processing module is connected with the data storage module, and the control module is respectively connected with the display screen and the image acquisition equipment;

2. The intelligent orthotic system that can perform knee joint functional assessment of claim 1, wherein: the display screen adopts an LCD liquid crystal touch screen, and the LCD liquid crystal touch screen supports a 4K display effect.

3. The intelligent orthotic system that can perform knee joint functional assessment of claim 1, wherein: the control module employs an English WEDAJetson TX2 core board.

4. The intelligent orthotic system that can perform knee joint function assessment according to any one of claims 1-3, wherein: the intelligent correction system capable of performing knee joint function evaluation further comprises a floor rubber mat, the evaluation and correction all-in-one machine and the image acquisition equipment are fixedly arranged above the floor rubber mat, and the evaluation and correction all-in-one machine and the image acquisition equipment are perpendicular to the floor rubber mat.